Quantitative Prediction Of Freshness Index Of Larimichthys Crocea And Research On Quality Traceability Technology

The Larimichthys crocea has become a popular seafood product because of its fresh and tender muscles and rich in nutrients such as protein,amino acids and fatty acids.There are many factors that affect the quality and quality of Larimichthys crocea.For example,Larimichthys crocea is prone to nutrient loss and spoilage during storage,Larimichthys crocea from different origins also have great differences in sensory quality and market price,it is more useful for other fish,It’s a way of pretending to be a Larimichthys crocea.Therefore,how to quickly measure the freshness index of Larimichthys crocea and at the same time realize the traceability and differentiation of Larimichthys crocea in terms of origin and species is of great significance to the research on the healthy development of Larimichthys crocea industry.In this paper,Fourier transform mid-infrared spectroscopy is used to establish a quantitative prediction model for the freshness of Larimichthys crocea during storage;Traceability technology for Larimichthys crocea origin.The main conclusions of the paper are as follows:(1)The basic nutrient composition data of Fujian cultured Larimichthys crocea,Wenzhou cultured Larimichthys crocea,Zhoushan wild Larimichthys crocea and Zhoushan cultured Larimichthys crocea were obtained,and one-way ANOVA(LSD and Tamhane T2 test)were carried out.The results showed that there were no significant differences in water,ash and crude protein content of Larimichthys crocea from the four production areas(P>0.05).The crude fat content of cultured Larimichthys crocea in Fujian was significantly higher than that of Larimichthys crocea in the other three producing areas(P<0.05).The content of Mg and K elements in Zhoushan wild Larimichthys crocea was the highest,and there was a significant difference between it and the other three kinds of cultured Larimichthys crocea(P<0.05).The Se element content of Zhoushan wild Larimichthys crocea was significantly higher than that of the other three types of cultured Larimichthys crocea.The content of Cr in the cultured Larimichthys crocea in Wenzhou was higher than that in Fujian(P<0.05).The content of Cu element cultured Larimichthys crocea in Wenzhou was the most,which was much higher than that in Fujian and Zhoushan wild Larimichthys crocea(P<0.05).The content of Zn element in Zhoushan wild Larimichthys crocea was higher than that in Fujian and Wenzhou cultured Larimichthys crocea(P<0.05).The contents of essential amino acids Thr,Ile,Leu,Phe,Lys and non-essential amino acids Asp,Glu,Pro and Ala in Zhoushan wild Larimichthys crocea were higher than those of the other three types of cultured Larimichthys croea(P<0.05).The essential amino acids Met,Leu,Phe,Lys,and the non-essential amino acids Ser,Glu,Ala,and Pro in the cultured Larimichthys crocea in Wenzhou were significantly different from those in the other three Larimichthys crocea(P<0.05).The content of non-essential amino acid Ser in the cultured Larimichthys crocea in Fujian was significantly different from that in the other three producing areas(P<0.05).The fatty acid content of C10:0 and C14:0 in Zhoushan Larimichthys crocea was significantly different from that of other three producing areas(P<0.05).The contents of fatty acids C16:0,C17:0,C17:1,C18:0,C18:1n9c,C20:3n3,C20:5n3,and C22:6n3 in Zhoushan wild Larimichthys crocea were significantly different from those of other three producing areas(P<0.05).The content of fatty acids C14:1,C20:0,C20:4n6,C22:1n9,and C24:1 of Larimichthys crocea cultured in Fujian was different from that of the other three producing areas(P<0.05).(2)Taking Zhoushan cultured Larimichthys crocea frozen at-18℃ for 0～150 days and at-45℃ for 0～160 days as the research object,the traditional physical and chemical analysis methods were used to analyze the freshness indexes(volatile salt base)during storage.TVB-N,TBAR,pH were measured,and Fourier transform infrared spectral characteristics were measured during the freezing period,and a quantitative analysis model was established by combining with TVB-N and TBAR.The results showed that the pH of cultured Larmichthys crocea showed a trend of first decreasing and then increasing under both storage conditions of-18℃ and-45℃,and the pH was 7.31 and 7.10 at the end of storage.The contents of TBAR and TVB-N both showed an upward trend,TVB-N was 21.91 mg/100g and 18.23 mg/100g at-18℃ and-45℃,and TBAR was 0.71mg MDA at-18℃ and 0.53mg MDA/100g at-45℃.By selecting spectral preprocessing methods such as baseline correction,SG smoothing,1st Der,2nd Der,MSC and SNV,and extracting spectral information at 3690～2800 cm-1 and 1800～650 cm-1 of characteristic waves.The results showed that at-18℃,a quantitative model was established with TVB-N,the optimal spectral preprocessing method was SNV,the optimal model was the partial least squares model(PLS)test set R2,and the RMSE was 0.839 and 1.267.The optimal spectral preprocessing method of establishing quantitative model with TBAR is 1st Der 5-point smoothing,and the optimal model is PLS model test set R2,RMSE is 0.871,0.088.At-45℃,the optimal preprocessing methods and models for the combination of TVB-N,TBAR and infrared spectra were MSC and PLS models,respectively.The results showed that the quantitative analysis model established by Fourier transform mid-infrared spectral data can realized the effective quantitative prediction of TBAR and TVB-N indicators reflecting fish freshness in Larimichthys crocea at-18℃ and-45℃.(3)Taking Fujian cultured Larimichthys crocea,Wenzhou cultured Larimichthys crocea,Zhoushan wild Larimichthys crocea and Zhoushan cultured Larimichthys crocea as research objects,a linear discriminant model(LDA)and a support vector machine model(SVM)based on Fourier transform mid-infrared spectroscopy were established to conduct the research on origin traceability.The results show that the use of the full spectrum at 4000～650 cm-1 through SG smoothing+3-point preprocessing to establish the SVM model has the best effect,and the accuracy rate of the test set is 83.3%.In the distinction between farmed Larimichthys crocea in Fujian and wild Larimichthys crocea in Zhoushan,the selected characteristic bands(3690～2800 cm-1+1800～650 cm-1)are the spectra after SNV,1st Der,2nd Der preprocessing and LDA modeling The accuracy of training set and prediction set is 100%.The characteristic band(3690～2800 cm1+1800～650 cm-1)after SNV and MSC pretreatment has the best effect of principal component analysis.After SNV pretreatment and cluster analysis,the Larimichthys crocea from the two origins are clustered into one category.(4)Based on the ratio of mineral elements and stable isotopes,the origin of Zhoushan wild Larimichthys crocea,Zhoushan cultured Larimichthys crocea,Fujian cultured Larimichthys crocea and Wenzhou cultured Larimichthys crocea were traced.Single factor for δ13C,δ15N,C,N,Na,Mg,Al,K,Ca,Fe,Ti,V,Cr,Mn,Co,Ni,Cu,Zn,Se,Sr,Mo,Sn,Ba elements one-way ANOVA(LSD and Tamhane T2 tests).The contents of Mg,Al,K,Ti,Cr,Cu,Zn,Se,Sn,Ba,δ13C,δ15N,C and N with significant differences is select for principal component analysis,cluster analysis and Fisher’s discriminant analysis.The results showed that the principal component analysis clearly distinguished the Larimichthys crocea from the four origins,of which the first two principal components accounted for 41.2%of PC1,PC2 accounted for 22.9%,and the first six principal components accounted for the cumulative contribution of variance was 90.5%.Cluster analysis showed that the large yellow croakers from the four origins were clustered into one group.Among the classification results of "leave one out" crossvalidation,85.7%of Zhoushan cultured Larimichthys crocea samples were correctly classified,14.3%of Zhoushan-cultured Larimichthys crocea were misclassified as Wenzhou cultured Larimichthys crocea,and the remaining three Larimichthys crocea samples were all classified as 100%correct identification.The results show that the application of mineral elements combined with stable isotopes can realize the origin traceability of Larimichthys crocea in different places.(5)The species identification of Larimichthys crocea,Larimichthys polyactis and Nibea albiflora was carried out by using Fourier transform infrared spectroscopy.The spectral data in the characteristic band 1800～650 cm-1 were preprocessed by 1st Der combined with SNV and 1st Der combined with MSC preprocessing,respectively.The two preprocessing methods were followed by principal component analysis to accurately distinguish the three fish species.The spectral data in the characteristic band of 1800～650 cm-1 was preprocessed by 1st Der and SNV,and cluster analysis found that the Larimichthys crocea,the Larimichthys polyactis and the Nibea albiflora were clustered into one group.The LDA model established by 1st Der,SNV and MSC pretreatment and the SVM model established by SNV and MSC pretreatment can distinguish between Larimichthys crocea,Larimichthys polyactis and Nibea albiflora.The results showed that Fourier transform infrared spectroscopy can be applied to qualitatively discriminate the Larimichthys crocea,the Larimichthys polyactis and the Nibea albiflora.